🎉 Super happy to announce that a significant part of @LavechinMarvin's and my PhD research on modelling early language acquisition has finally been published in Developmental Science!
📄 https://t.co/jLVSdM9FLd
(OS version : https://t.co/dyXmzj6zNi)
🧵1/7
🚀 Exciting Post-Doc Opportunity! 🚀
Join the CoML team for the new ERC project InfantSimulator ! If you're passionate about language modeling & machine learning, apply now ! @ENS_ULM
📍 Paris
🔗 https://t.co/C8LskSopb2
#PostDoc#CognitiveScience#LanguageModeling#Job#AI
[🚨Recruitment🗣️]
CoML team is actively recruiting a Postdoctoral Fellow with expertise in machine learning, linguistics, or cognitive science. More information in the detailed announcement here : https://t.co/jZeiWxupq9
🔉We just released EmphAssess, a benchmark for evaluating emphasis in speech to speech models.
This work is a product of my internship at @MetaAI, and we’d gladly appreciate any feedback !
📝 : https://t.co/3PTaxGxKOI
👩💻 : https://t.co/qhEyahnKuU
Hi network, trying my luck here! 🍀
I'm starting to look into post-PhD opportunities. If you know of any companies who could be interested in my profile (among others : speech modelling, SSL, evaluations, cognition), please drop their name! Or better yet, send them my profile 🙌
Accepted paper at Findings at #EMNLP2023 (with @itspablit0) on unsupervised speech segmentation into words. Our simple teacher-student method leverages XLS-R and improves over the latest SOTA by 130% on average over five different languages.
We have a second accepted paper at #EMNLP2023 (with @adiyossLC) on Generative Spoken Language Modelling (textless NLP). We managed to generate clean speech sentences using continuous units instead of the discrete HuBERT.
@CoML_ENS
[ Seminar 🗣️] This afternoon Sharon Goldwater gave a wonderful talk on Analyzing representations of self-supervised speech models and also perspective as a cognitive scientist on the possible connections with the study of humans' learning and representation of speech!
We are thrilled to welcome @HermanKamper, Karen Livescu and Sharon Goldwater to visit our Team!
[🗣️Seminar] This Monday @HermanKamper gave wonderful talk on Multimodal few-shot learning & probing self-supervised speech model.
@tallinzen returns to CoML after 6 years !🌟 His insightful talk on surprisal and syntactic disambiguation sheds light on the intriguing gap between humans and language models! Exciting research direction at the frontier between psycholinguistics 👶 and neural networks 🤖 !
Thrilled to announce our new paper with @LavechinMarvin & Emmanuel Dupoux, where we present our perspective on what it takes to build effective learning language simulations, now in Journal of Child Language! #psycholinguistics#computationalmodelling
https://t.co/FRIPX9eUsd
📢 Exciting news! We just released ProsAudit, a prosodic benchmark for SSL models of speech 🥳
💬 It is now part of the Zero Resource Speech Challenge (track 4). The paper also includes results on a human comparison. 👨💻🤖
📰Check out the preprint: https://t.co/gnMLi6KLBc
*Announcement!*
I'm actively looking for a postdoctoral position (anywhere in the world) starting in 2023 on speech and language processing for low-resourced languages (preferably endangered language documentation and description or automatic inference of typological features)!
Finally! 🎉⏰🎉
@LavechinMarvin and I just released the preprint of our latest paper "Statistical learning bootstraps early language acquisition" (🧵coming soon).
📢Psycholinguists and ML enthusiasts, check it out and let us know what you think!
https://t.co/IM9CfyfH38
🤝🎷 The lab will be represented at #NeurIPS2022 in New Orleans this year! Don't forget to go by @Mathieu_Rita's poster at the Poster Session 5 on Thu 1 Dec (Hall J #713) ! Thrilling chats coming !
🎷 I'll attend @NeurIPSConf this week to present our work "Emergent Communication: Generalization and Overfitting in Lewis Games" ! If you're interested, drop by the Poster Session 5 on Thu 1 Dec - Hall J #713.
🗞️Paper: https://t.co/KkAhSS1dah
🧵Thread: https://t.co/KqI3H8Bopq
The ALMAnaCH lab (@InriaParisNLP) at @inria_paris is looking for candidates for full-time permanent researchers in natural language processing (NLP) and digital humanities (DH) to join us in the beautiful city of Paris. We welcome profiles for both junior & senior positions. 1/5